[unreadable] Understanding how a child comes to identify individual words within fluent speech is crucial to understanding first language development. Fundamentally, language learning requires both recognition and segmentation skills. If infants cannot break input utterances into constituent components and determine whether the resulting auditory objects are exemplars of particular linguistic types, it will be impossible for them to learn what individual components mean and how these parts fit together. The representational units (e.g., words) serve as the basis upon which infants develop syntactic and semantic knowledge. By coupling a traditional infant speech perception procedure with an established form of optical neuroimaging, the research proposed in this application will provide researchers with a new tool for interpreting behavioral data from very young infants. The objective of this application is to identify neural correlates of lexical representations in infants of different ages using Near-Infrared Spectroscopy (NIRS). Groups of infants--aged 5.5 months and 10 months--will be tested using a modified preferential listening procedure while wearing a set of custom optical imaging probes and detectors on their heads. The specific aims of this study are to (1) construct localized maps of neural activity for groups of infants displaying familiarity and novelty listening preferences, and (2) differentiate between semantic and episodic auditory memory in young infants. The central hypothesis for the proposed research is that optical neuroimaging will reveal that infants have more nuanced representations of words than behaviorally conveyed familiarity and novelty effects can demonstrate. The outcome of these studies will be used to advance theories of infant language learning. These theories will further our understanding of both normal and abnormal language development. [unreadable] [unreadable]